[CI] Refactor pipeline and test RayCluster sample yamls #1321
Conversation
Signed-off-by: Archit Kulkarni <architkulkarni@users.noreply.github.com>
Signed-off-by: Archit Kulkarni <architkulkarni@users.noreply.github.com>
Signed-off-by: Archit Kulkarni <architkulkarni@users.noreply.github.com>
Signed-off-by: Archit Kulkarni <architkulkarni@users.noreply.github.com>
Signed-off-by: Archit Kulkarni <architkulkarni@users.noreply.github.com>
Signed-off-by: Archit Kulkarni <architkulkarni@users.noreply.github.com>
| # Install requirements | ||
| pip install -r tests/framework/config/requirements.txt | ||
|
|
||
| # Bypass Git's ownership check due to unconventional user IDs in Docker containers |
There was a problem hiding this comment.
This is the only new part of the env setup that was needed for the RayCluster test to work. It's required because the RayCluster test uses some git commands, but the owner of workdir was "2000" which git deems suspicious and throws an error:
stderr: 'fatal: detected dubious ownership in repository at '/workdir'
I assume this is somehow related to docker and that it's safe to ignore this.
.buildkite/setup-env.sh
Outdated
| install -o root -g root -m 0755 kubectl /usr/local/bin/kubectl | ||
|
|
||
| # Create the cluster | ||
| time kind create cluster --wait 120s --config tests/framework/config/kind-config-buildkite.yml |
There was a problem hiding this comment.
Why do we need to create a Kind cluster (L24 - L56) here without running any tests?
There was a problem hiding this comment.
We don't need it anymore, this is just from the initial test. I have a followup PR which will remove it
There was a problem hiding this comment.
I just removed it in this PR.
.buildkite/setup-env.sh
Outdated
|
|
||
| # Install python 3.10 and pip | ||
| apt-get update | ||
| apt-get install -y python3.10 python3-pip |
There was a problem hiding this comment.
We may update this based on #1325.
There was a problem hiding this comment.
Good eye! If your PR gets merged first I'll fix the merge conflict in this PR
There was a problem hiding this comment.
#1325 has already been merged.
There was a problem hiding this comment.
O you're right
Signed-off-by: Archit Kulkarni <architkulkarni@users.noreply.github.com>
…move-raycluster-sample-yaml-test Signed-off-by: Archit Kulkarni <architkulkarni@users.noreply.github.com>
Signed-off-by: Archit Kulkarni <architkulkarni@users.noreply.github.com>
…move-raycluster-sample-yaml-test
Signed-off-by: Archit Kulkarni <architkulkarni@users.noreply.github.com>
Signed-off-by: Archit Kulkarni <architkulkarni@users.noreply.github.com>
Signed-off-by: Archit Kulkarni <architkulkarni@users.noreply.github.com>
Signed-off-by: Archit Kulkarni <architkulkarni@users.noreply.github.com>
|
@kevin85421 all comments are addressed and I expect tests to pass, can you take another look? |
…1321) Refactor pipeline and test RayCluster sample yamls
…1321) Refactor pipeline and test RayCluster sample yamls
Why are these changes needed?
Moves the sample YAML tests from github actions to Buildkite.
This PR now tests 10 sample YAML files, instead of only three as before. We omit the TPU sample YAML because we don't have TPUs on buildkite.
Related issue number
Checks